Applying the Parallel GPU Model to Radiation Therapy Treatment
Department of Computer Science, University of Texas at Dallas, Richardson, TX USA
The 2013 International Conference on Parallel and Distributed Processing Techniques and Applications (PDPTA’13), 2013
@article{kirtzic2013applying,
title={Applying the Parallel GPU Model to Radiation Therapy Treatment},
author={Kirtzic, J Steven and Allen, David and Daescu, Ovidiu and Richardson, TX},
year={2013}
}
With current advances in high performance computing, particularly the applications of GPUs, it is easy to see the need for a model for GPU algorithm development. We developed a model which offers a multi-grained approach intended to accommodate nearly any GPU. Radiation therapy is one of the most effective forms of cancer treatment available. In order to minimize the risk to the patient, physicians design treatment plans that expose the tumor to the prescribed levels of radiation while minimizing the exposure to the surrounding tissues. Our system allows users to quickly and easily visualize and compare treatment plans in order to identify the best one, with the most critical aspect of the simulation being implemented on the GPU using our parallel algorithm design model. In this paper, we show how the application of our model results in significant increases in algorithm performance, particularly in radiation therapy treatment simulation.
January 5, 2014 by hgpu
